KeenKT: Knowledge Mastery-State Disambiguation for Knowledge Tracing

TL;DR

KeenKT introduces a probabilistic approach to knowledge tracing that models student knowledge states as distributions, capturing learning fluctuations and improving prediction accuracy over existing methods.

Contribution

The paper proposes KeenKT, a novel knowledge tracing model using distributional state representations and attention mechanisms to better handle student behavior variability.

Findings

Outperforms state-of-the-art models in accuracy and sensitivity.

Achieves up to 5.85% AUC and 6.89% ACC improvements.

Demonstrates robustness across six datasets.

Abstract

Knowledge Tracing (KT) aims to dynamically model a student's mastery of knowledge concepts based on their historical learning interactions. Most current methods rely on single-point estimates, which cannot distinguish true ability from outburst or carelessness, creating ambiguity in judging mastery. To address this issue, we propose a Knowledge Mastery-State Disambiguation for Knowledge Tracing model (KeenKT), which represents a student's knowledge state at each interaction using a Normal-Inverse-Gaussian (NIG) distribution, thereby capturing the fluctuations in student learning behaviors. Furthermore, we design an NIG-distance-based attention mechanism to model the dynamic evolution of the knowledge state. In addition, we introduce a diffusion-based denoising reconstruction loss and a distributional contrastive learning loss to enhance the model's robustness. Extensive experiments on six public datasets demonstrate that KeenKT outperforms SOTA KT models in terms of prediction accuracy and sensitivity to behavioral fluctuations. The proposed method yields the maximum AUC improvement of 5.85% and the maximum ACC improvement of 6.89%.